ミスチルで一番使われている単語は「君」と「僕」

import pandas as pd
import collections
from janome.analyzer import Analyzer
from janome.charfilter import *
from janome.tokenfilter import *
import plotly.graph_objects as go
# csvファイルを読み込み
df_file = pd.read_csv('mrchildren-lyrics.csv', encoding='utf-8')
# 歌詞をデータフレームからリストに変換
song_lyrics = df_file[['曲名', '歌詞']].values.tolist()
results1 = []
results2 = []
results3 = []
results4 = []
for s in song_lyrics: # 複合名詞化 a = Analyzer(token_filters=[CompoundNounFilter()]) result = [token.base_form for token in a.analyze(s[1]) if token.part_of_speech.split(',')[0] in ['名詞']] # 単語の頻度を確認 c = collections.Counter(result) result_list1 = [s[0], c['君']] result_list2 = [s[0], c['僕']] result_list3 = [s[0], c['君'] + c['僕']] results1.append(result_list1) results2.append(result_list2) results3.append(result_list3) if (c['君'] + c['僕']) > 10: result_list4 = [s[0], c['君'] - c['僕']] results4.append(result_list4)
# 「君」の使用数を曲ごとにグラフ化
results1 = dict(results1)
graph_list1 = {k:v for k, v in results1.items() if v > 0}
# 君が使われている曲数を確認
print(len(graph_list1))
# plotlyでグラフ化
fig = go.Figure([go.Bar(x=list(graph_list1.keys()), y=list(graph_list1.values()))])
fig.update_layout(xaxis={'categoryorder':'total descending'})
fig.show()
# グラフをhtml形式で保存
filename = "kimi"
with open(filename + '.html', 'w') as f: f.write(fig.to_html(include_plotlyjs='cdn'))
# 「僕」の使用数を曲ごとにグラフ化
results2 = dict(results2)
graph_list2 = {k:v for k, v in results2.items() if v > 0}
# 僕が使われている曲数を確認
print(len(graph_list2))
# plotlyでグラフ化
fig = go.Figure([go.Bar(x=list(graph_list2.keys()), y=list(graph_list2.values()))])
fig.update_layout(xaxis={'categoryorder':'total descending'})
fig.show()
# グラフをhtml形式で保存
filename = "boku"
with open(filename + '.html', 'w') as f: f.write(fig.to_html(include_plotlyjs='cdn'))
# 「君」+「僕」+「君+僕」をグラフ化
results3 = dict(results3)
graph_list3 = results3
fig = go.Figure(data=[ go.Bar(x=list(graph_list1.keys()), y=list(graph_list1.values()), name='君'), go.Bar(x=list(graph_list2.keys()), y=list(graph_list2.values()), name='僕'), go.Bar(x=list(graph_list3.keys()), y=list(graph_list3.values()), name='君＋僕')
])
fig.update_layout(barmode='group')
fig.update_layout(xaxis={'categoryorder':'total descending'})
fig.show()
# グラフをhtml形式で保存
filename = "kimi-and-boku"
with open(filename + '.html', 'w') as f: f.write(fig.to_html(include_plotlyjs='cdn'))
# 「君」 - 「僕」をグラフ化
results4 = dict(results4)
graph_list4 = results4
print(len(graph_list4))
# plotlyでグラフ化
fig = go.Figure([go.Bar(x=list(graph_list4.keys()), y=list(graph_list4.values()))])
fig.update_layout(xaxis={'categoryorder':'total descending'})
fig.show()
# グラフをhtml形式で保存
filename = "kimi-boku"
with open(filename + '.html', 'w') as f: f.write(fig.to_html(include_plotlyjs='cdn'))